The number of social media platforms with which users interact has proliferated over the past few years. Examples of such social media platforms include social networking systems (e.g., Facebook), professional networking systems (e.g., LinkedIn), virtual world platforms (e.g., Second Life), messaging systems (e.g., Google email (Gmail), Google Wave, Skype), blogging systems (e.g., Blogspot.com), and review/rating systems (e.g., Yelp.com, Digg.com). Social networking platforms, such as Facebook, are continuing to gain popularity as platforms on which users interact, communicate and share using multiple types of data and communication channels. For example, a number of social networking platforms provide one or more messaging tools, and photo and video sharing capabilities. These social networking platforms also allow users to share content located on the Internet with each other in a convenient manner, and provide mechanisms by which users can exercise control over with whom they share information, and what information is shared with them (e.g., by source or by content type).
Virtual worlds similarly host vibrant communities of people who interact, play, do business and even find romance online.
This proliferation has presented a number of challenges to users, both from a resource perspective (e.g., the time required to manage interactions across multiple social media platforms) and technical perspective (e.g., having to learn user interfaces and privacy controls across multiple platforms). For example, managing interactions across multiple social media platforms may require a user to duplicate actions (e.g., the publication of updates or other information). A particular user may also be presented with duplicate information from other users, via a number of social media platforms (e.g., when a user publishes the same information on both Facebook and MySpace). Reacting to such communications and events across multiple social media platforms may require more time than a user is willing or able to expend on social networking activities.
Consider also that privacy controls across multiple social media platforms may vary substantially. It is burdensome for a user to have to learn and master various types and flavors of privacy controls that are provided by multiple social media platforms. Indeed, it is not uncommon for a user, as a result of a lack of understanding of privacy controls, to have a certain of their information published to unintended recipients via a social media platform. Particularly as the complexity of social media platforms has increased (e.g., as a result of the opening of such platforms to third-party developers and applications), the challenge of exercising a desired degree of privacy control across multiple platforms has become daunting to many users.
Notwithstanding the challenges presented above, many users desire to maintain an active presence on their social platforms, and to be actively engaged with their social networks on a regular basis.
Hattori et al., in their paper entitled “Socialware: Multiagent Systems for Supporting Network Communities”, discussed their development of multi-agent systems to assist in various social activities on network communities, which they term “socialware”. Hattori et al. described a network community as a collection of personal units, community agent(s) and a set of relationships between them. A personal unit is described as consisting of a user and his or her personal agent. Each personal agent can help the user by gathering and exchanging information, visualizing contexts, and recommending or assisting the user in making a particular choice. The personal agents of a user may cooperate and act as a unit, with the user being the central figure. The community of agents has the function of providing shared information, knowledge or context within a community, and act as mediators for informal communications between people. An architecture where each user has personal agents that communicate with each other enables the community to spread. Hattori et al. states that it is possible to have some agents be domain-specific (for example, an information retrieval agent specialized for financial news) and others to be more generic (for example, an interface agent for navigating and reading documents).
Adoption of multiple aspects of a user and the user's changes in interests can be achieved by changing the system dynamically and autonomously. For example, a domain-specific agent can clone itself and produce a new agent that makes additional communication channels when the user's interest has changed.
Hattori et al. furthermore described the development of a prototype application for the purposes of linking people (a “CommunityOrganizer”), which consists of a personal agent for each user, and a community agent. Each personal agent functions to acquire the user profile and to visualize potential communities around a user. The community agent functions to collect the user profiles, and to maintain the information on potential communities. The relevancies between users are calculated by the community agent from the users' profile data. These profiles can be obtained from each user's input, from archives of mailing lists using keyword extraction techniques, or from user information on the Web. Each personal agent is furthermore described as having slide bars which temporarily adjust the weightings of the viewpoints, since a degree of common interest consists of multiples aspects. Each personal agent displays structures of discussions according to the user's interests. The community agent may have to classify messages according to several criteria, such as topic, time and reputation.
Partsakoulakis et al., in their paper entitled “Representative Agents for Reliable Participation in Social Contexts”, describe a prototype system geared towards empowering humans to deliberately form and manage their social context and position via personal agents that act as their representatives. Personal agents are described as representing humans, and form their “digital analog” within organizations. The approach is based on a role-based model concerning consistency and reliability of role playing within a social context. Agents are described as being aware of an overall social context, profile and needs of their users, and may search for, evaluate (e.g., in terms of consistency), and present relevant information to their users. Personal agents serve/represent specific humans and populate groups that play organizational roles. Personal agents are described as maintaining the profiles of the humans that they represent (e.g., their preferences, their roles, etc.) to help them achieve their goals. Personal agents can have managerial responsibilities within an organization, and their architecture is described as comprising a knowledge base, an inference engine, and an interface of the agent with its environment. Two types of interfaces are distinguished, namely the interface between the agent and the user, and the interface between the agent and other agents. The knowledge base of personal agent is described as comprising the social context in which the agent operates, the profile of the user, and a cache of addresses of other agents.